import torch 

import torch.nn as nn 

class character_cnn(nn.Module):

    def __init__(self, vocabulary, sequence_length, number_classes = 10):

        super().__init__()

        self.conv1 = nn.Sequential(nn.Conv1d(len(vocabulary)+1, 256, kernel_size = 7, padding = 0),

                                   nn.ReLU(),

                                   nn.MaxPool1d(3)

                                   )

        self.conv2 = nn.Sequential(nn.Conv1d(256, 256, kernel_size=7, padding=0),

                                   nn.ReLU(),

                                   nn.MaxPool1d(3)

                                   )

        self.conv3 = nn.Sequential(nn.Conv1d(256, 256, kernel_size=3, padding=0),

                                   nn.ReLU()

                                   )

        self.conv4 = nn.Sequential(nn.Conv1d(256, 256, kernel_size=3, padding=0),

                                   nn.ReLU()

                                   )

        input_shape = (1, len(vocabulary)+1, sequence_length)

        self.output_dimension = self._get_conv_output(input_shape)

        self.fc1 = nn.Sequential(

            nn.Linear(self.output_dimension, 1024),

            nn.ReLU(),

            nn.Dropout(0.5)

        )

        self.fc2 = nn.Sequential(

            nn.Linear(1024, 1024),

            nn.ReLU(),

            nn.Dropout(0.5)

        )

        self.fc3 = nn.Linear(1024, number_classes)

        self.act = nn.Sigmoid()

    def _get_conv_output(self, shape):

        x = torch.rand(shape)

        x = self.conv1(x)

        x = self.conv2(x)

        x = self.conv3(x)

        x = self.conv4(x)

        x = x.view(x.size(0), -1)

        output_dimension = x.size(1)

        return output_dimension

    def forward(self, x):

        x = self.conv1(x)

        x = self.conv2(x)

        x = self.conv3(x)

        x = self.conv4(x)

        x = x.view(x.size(0), -1)

        x = self.fc1(x)

        x = self.fc2(x)

        x = self.fc3(x)

        x = self.act(x)

        return x